Linked Art for Archaeological Data Exchange

Autor: Gruber, Ethan
Přispěvatelé: Smith, Tyler Jo, Gondek, Renee
Jazyk: angličtina
Rok vydání: 2021
Předmět:
DOI: 10.5281/zenodo.4956818
Popis: Linked Art (https://linked.art/) is an emerging community of museum and information science specialists working to establish a common profile for the exchange of cultural heritage data by means of current Linked Open Data (LOD) and Application Programming Interface (API) methodologies. Building on similar principles outlined by the International Image Interoperability Framework (IIIF: https://iiif.io/), Linked Art is a particular implementation of the CIDOC Conceptual Reference Model (CRM) rendered in the JSON-LD syntax. Linked Art aims to strike a balance between completeness and usability, and while not all possible permutations of human knowledge are intended to be represented in the Linked Art profile, the minimum level of detail, with respect to both the data model and controlled vocabulary integration, is greater than more generalized aggregation models, such as those used for Europeana (https://www.europeana.eu) or the Pelagios Network (http://commons.pelagios.org/). The aim of Linked Art is to achieve Linked Open Useable Data (LOUD), lowering the barriers for developers to reuse data for a variety of research contexts. Many of the member institutions of the Linked Art steering committee hail from fine arts museums, but there exists some overlap between several members and archaeological materials, particularly among encyclopedic museums and research institutes, such as the Getty Museum and Smithsonian. The American Numismatic Society is among the earliest of implementers of the Linked Art API specification, extending the open-source Numishare software framework for its online collection of coins and similar objects to output JSON-LD conforming to a numismatic model proposed by the author. In a similar vein, the Kerameikos.org project has aligned its aggregation model nearly completely with the Linked Art CIDOC-CRM profile. Kerameikos.org is a collaborative effort to define the intellectual concepts of ceramics studies following the principles of Linked Open Data, focusing initially in Archaic and Classical Greece (c. 700-330 BCE) (Gruber and Smith 2014, 205-214). The project is developing workflows to incorporate vase data from across archaeological and museum datasets in order to facilitate new types of query and visualization. This paper addresses recent developments in the harvesting and normalization of Linked Art-compliant JSON-LD data into both Kerameikos.org and Nomisma.org’s LOD ecosystems. Findspots for Greek vases implement a proposal to the Linked Art community based on the ARIADNE Plus model that extends CIDOC-CRM for archaeological context (PIN 2019; Gruber 2019). While many artifacts held by museums were not scientifically excavated and cataloged according to modern standards for data recording, the database records for these artifacts may still include very basic metadata corresponding to the place of finding and general context of the place (e.g., a tomb) or techniques used (metal detecting, as is the case for many coin finds). The prototype of this data harvesting system in Kerameikos.org was built primarily upon test data provided about several Greek vases held by the Indianapolis Museum of Art at Newfields; one of these has a findspot of Vulci, an Etruscan settlement in Italy, which would serve as a proof of concept for processing the Linked Art implementation of the ARIADNE archaeological data model. This paper details the intricacies of this workflow: 1) using XForms to request and process JSON-LD, 2) coreference concept URIs in the JSON-LD (typically, Getty vocabularies) with applicable Kerameikos.org ones, 3) normalize gazetteer URIs associated with findspots to Wikidata.org entities, and use Wikidata APIs to extract geographic coordinates and hierarchy (following the CIDOC-CRM spatiotemporal extension), and 4) serialize these data into CIDOC-CRM RDF/XML and post into the Kerameikos.org SPARQL endpoint. This is an extension of the data integration workflow developed in Nomisma.org in 2015, and the processes outlined in this paper may be of interest to others working to aggregate data in other sectors of cultural heritage. In conclusion, this paper is a demonstration of the development of a bridge between museum and archaeological data best practices. While Linked Art continues to evolve toward a version 1.0 specification, the application of ARIADNE’s principles for CIDOC-CRM representation of archaeological context will hopefully enable museums to align their digital collections to the research questions of researchers who are interested in the patterns of distribution of materials over time and space. Likewise, the JSON-LD oriented approach to the Linked Art API may serve as a use case to the archaeological community to promote optimal data reuse through increasingly common serializations of Linked Open Data. References Gruber, Ethan and Tyler Jo Smith. 2014. “Linked Open Greek Pottery.” In Proceedings of the Computer Applications in Archaeology Conference 2014, edited by F. Giligny, et al (Oxford: Archaeopress, 2015), 205-214. https://doi.org/10.5281/zenodo.258046. Gruber, Ethan. 2019. “ARIADNE-compliant archaeological find model.” Linked.art Github issue. Accessed October 2019. https://github.com/linked-art/linked.art/issues/285. PIN, University of Florence. 2019. “Definition of the CRMarchaeo, Version 1.4.8, February 2019”. Accessed October 29, 2019. http://www.cidoc-crm.org/crmarchaeo/sites/default/files/CRMarchaeo_v1.4.8.pdf.
{"references":["Gruber, Ethan and Tyler Jo Smith (2004). Linked Open Greek Pottery. 10.5281/zenodo.258046"]}
Databáze: OpenAIRE
Externí odkaz: